Recombinant adeno-associated virus (AAV) vectors have emerged as promising vehicles for gene transfer as they are no-pathogenic, have a wide host range, and transduce non-proliferating cells. Furthermore, wild type AAV possesses the unique property of integration into a specific site in human chromosome 19 termed AAVS1, a property that has been mapped to the AAV encoded Rep protein. However, for reasons outlined in the proposal, Rep has been removed from all commonly used rAAV vectors. We and others have demonstrated that wild type AAV free, rep negative rAAV vectors integrate into primary human cells but not into AAVS1. Furthermore, Southern analyses have demonstrated that these vectors do not insert randomly, but may have preferred integration sites in primary human cells. Finally, results from other laboratories indicate the frequency of integration issues which are germane to the further development of rAAV vectors for clinical gene therapy. To gain insight about the biology of rep negative rAAV vector integration, inverse PCR, shuttle vectors, and lambda phage cloning will be employed to identify, isolate, and sequence the cellular-vector junction fragments of vectors integrated into primary human cells. In addition, we will develop methods, including use of Rep-linked rAAV virions and fusion of Rep to a herpes virus tegument protein (VP22) which has been shown to freely traffic between cells and accumulate in the nucleus, to provide Rep gene functions in trans in an attempt to both increase the frequence of vector integration, and to promote site specific integration.